Standard deviation change - negative values

MRCIEU / PHESANT

PHESANT - PHEnome Scan ANalysis Tool (pheWAS, Mendelian randomisation (MR)-pheWAS etc.) in UK Biobank

MIT License

136 stars 69 forks source link

Standard deviation change - negative values #16

Closed C-Gouveia closed 4 years ago

C-Gouveia commented 4 years ago

Hi! I'm facing now another issue. I'm not sure how to interpret negative values for "Standard deviation change" - I have some in the forest plots for continuous data. Could you explain me very briefly? Sorry for this, and thank you very much!

louise commented 4 years ago

Continuous outcomes are inverse rank normal transformed by PHESANT. Therefore an association of e.g. -1 would mean:

Each unit increase of your exposure variable, is associated with a 1 SD lower inverse rank normal transformed outcome variable.

If you are using the standardise=TRUE PHESANT argument (which is default), then it means: Each SD increase of your exposure variable, is associated with a 1 SD lower inverse rank normal transformed outcome variable.